Hi committers,

I’ve noticed a few commits going into Apache lately that are not following the 
format of [AMBARI-XXXX] – The title of my Jira

This makes it difficult to correlate a commit to the actual Jira which it was 
for. As an example of the current git log (red text has no Jira):

75489756dd [AMBARI-23993] Mpack Instance Manager should create pid dir and log 
dir for each instance (#1424)
6a531ca06c Clear clusterData when user is not logged in.
3d92d03c0d AMBARI-23987. Set 'skip.service.checks' = true during deploy.
3dcef6ec91 AMBARI-23987. Set 'skip.service.checks' = true during deploy.
f8293cff7b AMBARI-23746. Use List for componentID duplicate check
9120330456 AMBARI-23746. Cannot query services with same name
85c6c6e6cc AMBARI-22875. Service group name mismatch
096dff5eeb AMBARI-23746. Cannot create same named component in different 
service groups in same request
0e9317e7e7 AMBARI-23746. Cannot create same named service in different service 
groups in same request
938be76315 AMBARI-14714. Fix ServiceResourceProviderTest
34900292d4 AMBARI-14714. Fix some unit tests
dedc8c424f Remove references to deleted unit test files and fix broken tests.
2d17eb4c4f ClusterStatusReport creates a map of host_id to component_name, 
needed change for resolving duplicate key

Please make sure you are following the guidelines for commits specified at: 
https://cwiki.apache.org/confluence/display/AMBARI/How+to+Contribute

In general, you can follow these simple rules:

  *   Every commit should be of the form [AMBARI-XXXX] followed by the Jira 
title
  *   Always use Squash and Merge with regular pull requests. There should be 
exactly 1 commit for every Jira, and this will ensure that updates to your PR 
are done in a single commit
  *   For pull requests which are syncing branches, you can use the Rebase and 
Merge option as long as there are no merge commits or conflicts
  *   If you are syncing branches and there are merge commits, you’ll need to 
use the Create Merge Commit option


Reply via email to